1. Field of the Invention
The present disclosure relates to an imaging apparatus having an object detection function.
2. Description of the Related Art
Japanese Patent Application Laid-Open No. 2010-96963 discusses a video camera system that includes a configuration for detecting a person's face from a captured video image and a configuration for tracking a predetermined object included in the video image by using pattern matching.
Japanese Patent Application Laid-Open No. 2008-5110 discusses a single-lens reflex camera that includes a second image sensor in addition to an image sensor intended for image capturing. Light transmitted through a focusing screen for an object image to be formed on and a display unit for displaying a selected focusing frame is incident on the second image sensor. In the single-lens reflex camera discussed in Japanese Patent Application Laid-Open No. 2008-5110, the second image sensor generates a video signal and the acquired video signal can be processed to track a predetermined object.
The foregoing techniques can be combined into the configuration that includes a second image sensor on which light transmitted through a focusing screen and a display unit for displaying a focusing frame is incident, and that detects a person's face from a video signal obtained by the second image sensor.
Since the light incident on the second image sensor has been transmitted through the display unit for displaying a focusing frame, the video image obtained from the second image sensor also includes a display of a selected focusing frame. If the focusing frame overlaps a person's face, the face is partially hidden and may become incapable of precise detection.
FIGS. 11A and 11B are diagrams illustrating a problem that occurs when a displayed focusing frame hides a part of a face. The camera can detect the faces of the two persons by using a video signal obtained from the second image sensor in a phase where no focusing frame is displayed.
If a plurality of faces is detected, the camera determines an evaluation value of each of the faces based on the size of the face, the position of the face, and the degree of reliability of the face. The camera then selects a face of the highest evaluation value as a main object. The degree of reliability of a face refers to a value that indicates the probability of an area being a face. The camera determines an area where the probability exceeds a predetermined threshold to be a face area.
The camera determines the right face to be a main object because the right face is greater in size, is closer to the image plane center, and has a higher degree of reliability of a face than the left face. As illustrated in FIG. 11A, the camera selects a focusing frame that lies in a position overlapping the right face, and displays the focusing frame in color.
The focusing frame in color makes the right face undetectable or significantly reduces the degree of reliability of the right face. As a result, the camera may determine the left face to be a main object as illustrated in FIG. 11B.
In other words, if there is a plurality of faces and a focusing frame is displayed over a face that is selected as a main object, it becomes difficult to select the same face to be a main object again.
The foregoing description has been given in the context of the function of detecting a face as an object. A similar problem can occur if a camera has a function of detecting a predetermined particular object (for example, a particular person, or a particular animal or physical object like a dog and a car).